Preeclampsia occurs in 7-10% of pregnancies causing extensive fetal and maternal mortality and morbidity. Its etiology is obscure, and therapy empiric. In this proposal we will examine the abnormal trophoblastic invasion of decidual vessels and its linkage to systemic manifestations of the disorder and test the hypothesis that trophoblast histocompatibility antigens are differently expressed and regulated in preeclampsia. Predictive tests will also be established and validated. We plan to characterize the determinants of the normal interaction of early human trophoblast with extracellular matrix in vitro and examine the expression and regulation of HLA antigens in trophoblast. To determine whether these changes are primary, similar studies performed on first trimester chorionic villus samples will be matched with pregnancy outcome. Animal and human studies will test the hypothesis that endothelial injury causes many of the pathophysiological changes of preeclampsia. Evidence for endothelial cell injury in vivo will be determined by examination of maternal and umbilical vessels and by quantitating endothelial fibronectin and mitogenic activity in the blood of preeclamptic women. In preliminary studies, sera from preeclamptic women, prior to but not after delivery, caused cytolysis of endothelial cells in vitro. We will extend these studies and determine their relevance to the pathophysiology of the disorder by examining endothelial cell functions including procoagulant activity, prostacyclin production and vasodepressor activity, and expression of platelet derived growth factor after exposure to preeclamptic and control serum. The factor(s) in serum leading to these changes will be characterized and their source determined. Our hypothesis, that the origin is from trophoblast and is responsible for the multisystemic changes of preeclampsia, will be specifically tested. Sera from preeclamptic women and matched controls, obtained and stored from early pregnancy, will be analyzed for the serum factor and fibronectin to test that these changes precede clinically evident preeclampsia; and correlated with responses of skin blood flow to non-invasive vasoconstrictor and dilatory stimuli to establish predictive tests for preeclampsia. An animal model of preeclampsia, pregnant rhesus monkeys made hypertensive and proteinuric by partial aortic occlusion, will be used to determine whether endothelial injury is the result of reduced vascular perfusion or a primary pathogenetic factor. Serum from these animals will be assayed for fibronectin, mitogenic and endothelial cell cytolytic activity before and at intervals after aortic narrowing and correlated with hemodynamic changes. Examination of the monkey's decidua and placentas will determine if differences found in these tissues are the cause or result of reduced trophoblast perfusion. Likewise HLA antigen shedding, proposed as a causal factor in the altered HLA expression, will be examined in this model to test its possible primary role. With these studies we hope to identify preeclamptic patients at an early stage, better understand the etiology of the disorder and provide insights which aid rational therapeutic approaches.